Decision Tree Regressor Simplifies with Limited Data

Want to predict the future with data? Start with Linear Regression. 📊 Linear Regression is one of the most powerful tools in Data Science. And it's simpler than it sounds. Here's what it does in plain English: It finds the relationship between two variables and uses it to make predictions. Real life example: → Hours studied vs Exam score → More hours studied = higher score → Linear Regression draws that line. 📐 In Python? Just 3 lines: from sklearn.linear_model import LinearRegression model = LinearRegression() model.fit(X, y) That's it. Your model is predicting. 🤯 Descriptive analytics tells you WHAT happened. Linear Regression tells you WHAT WILL happen. And that's where data gets exciting. 🚀 Initial stage, making predictions might seems difficult and confusing but with constant practice, it turn fun.. what it takes to be a data analyst.... #DataScience #LinearRegression #Python #DescriptiveAnalytics #MachineLearning #BeginnerCoder #LearnToCode

🚀 Day 38 of My 90-Day Data Science Challenge Today I worked on Cross-Validation (K-Fold Validation). 📊 Business Question: How can we evaluate a machine learning model more reliably using all available data? Instead of a single train-test split, Cross-Validation divides data into multiple parts and evaluates the model multiple times. Using Python & scikit-learn: • Applied KFold / cross_val_score() • Split dataset into multiple folds • Trained model on different subsets • Evaluated performance across folds • Calculated average accuracy 📈 Key Understanding: Each data point gets a chance to be used for both training and testing. 💡 Insight: Cross-validation reduces the risk of overfitting and gives a more stable performance estimate. 🎯 Takeaway: Evaluating models multiple times leads to more reliable and trustworthy results. Day 38 complete ✅ Improving model evaluation techniques 🚀 #DataScience #MachineLearning #CrossValidation #Python #LearningInPublic #90DaysChallenge

🚀 Day 43 of My 90-Day Data Science Challenge Today I worked on Stacking (Ensemble Learning Advanced Technique). 📊 Business Question: How can we combine multiple different models to improve prediction performance? Stacking is an ensemble technique where multiple models are combined and a meta-model learns from their predictions. Using Python & scikit-learn: • Learned concept of base models & meta-model • Combined predictions from multiple models • Applied stacking technique • Improved overall model performance • Compared with individual models 📈 Key Understanding: Stacking uses outputs of multiple models as input to a final model for better predictions. 💡 Insight: Different models capture different patterns — combining them improves accuracy. 🎯 Takeaway: Smart model combinations can outperform even the best individual model. Day 43 complete ✅ Exploring advanced ensemble techniques 🚀 #DataScience #MachineLearning #Stacking #EnsembleLearning #Python #LearningInPublic #90DaysChallenge

Day 51 of my #100DaysOfCode challenge 🚀 Today I worked on a Python program to perform Matrix Transpose using NumPy. This is a fundamental concept in linear algebra and widely used in Data Science & Machine Learning. What the program does: • Creates a 2D matrix using NumPy • Transposes the matrix (rows ↔ columns) • Uses built-in .T for efficient computation • Displays original and transposed matrix Original Matrix: [1, 2, 3] [4, 5, 6] [7, 8, 9] Transposed Matrix: [1, 4, 7] [2, 5, 8] [3, 6, 9] How the logic works: • Create matrix using NumPy array • Use: 👉 matrix.T • This automatically swaps: • Rows → Columns • Columns → Rows • No manual loops required ✅ Why this is important: – Core concept in Linear Algebra – Used in Machine Learning algorithms – Essential for matrix operations & transformations – Makes code faster and cleaner with NumPy – Time Complexity: O(n × m) – Space Complexity: O(1) (view-based operation) Key learnings from Day 51: – Introduction to NumPy – Matrix transpose concept – Efficient built-in operations – Writing optimized Python code #100DaysOfCode #Day51 #Python #NumPy #DataScience #MachineLearning #Matrix #LinearAlgebra #CodingPractice #ProblemSolving #DeveloperJourney #BuildInPublic #BTech #CSE #AIandML #VITBhopal #TechJourney

🚀 Day 3 – #Daily_DataScience_Code Taking the next step in our data science journey 👩💻 Today, we move beyond CSV files and explore how to read Excel files with multiple sheets 📊 💻 What we did today: - Loaded an Excel file directly from the web 🌐 - Read all sheets at once using pandas - Retrieved available sheet names - Accessed a specific sheet using its name (not index) - Displayed the first rows using head() 🎯 Key Insight: When working with Excel files, using sheet names makes your code more robust and readable, especially when dealing with multiple datasets. Let’s keep building step by step 🚀 #DataScience #MachineLearning #Python #AI #DataHandling #LearnByDoing #DataScienceWithDrGehad #DailyDataScienceCode

Pandas just made a lot more sense to me. Spent the last few days working through data manipulation in pandas and honestly it clicked better than I expected. Here is what I covered. Filtering rows means pulling only the data you actually need, like WHERE in SQL but in Python. Adding and dropping columns helped me clean up messy datasets and that part felt really satisfying. GroupBy is basically pivot tables in Excel but way more flexible. And handling missing values because real world data is never clean. The part that surprised me was how much you can do in just 2 or 3 lines of code. What used to feel like a lot of steps in Excel just happens. Still getting comfortable with chaining multiple operations together. That part is a bit tricky but I am getting there. If you are learning pandas too drop a comment. Would love to swap notes. #DataScienceJourney #LearningInPublic #Pandas #Python #DataAnalytics #100DaysOfCode #DataScience #MachineLearning #AI

🚀 Day 54 of My 90-Day Data Science Challenge Today I worked on Loss Functions in Machine Learning. 📊 Business Question: How do we measure how wrong a model’s predictions are? Loss functions calculate the difference between actual and predicted values. Using Python concepts: • Learned Mean Squared Error (MSE) • Understood Mean Absolute Error (MAE) • Explored Log Loss (Binary Cross-Entropy) • Compared regression vs classification loss • Understood impact on model training 📈 Key Understanding: Loss functions guide the model to improve by minimizing error. 💡 Insight: Choosing the right loss function is crucial for correct model learning. 🎯 Takeaway: Better loss function → better learning → better predictions. Day 54 complete ✅ Understanding model errors 🚀 #DataScience #MachineLearning #DeepLearning #LossFunction #Python #LearningInPublic #90DaysChallenge

Can a simple chart reveal financial trends? 📈 While exploring the dataset, I wanted to compare how key financial indicators evolved over time. Instead of looking at the numbers in tables, I transformed the data and visualized it using a bar plot in Python. I used Pandas, Seaborn, and Matplotlib to reshape the dataset and plot the values of Equity Capital, Reserves, and Total Assets across different years. 📈 What this visualization helps show: • How financial indicators change year by year • The relative scale between Equity Capital, Reserves, and Total Assets • Overall growth patterns that are harder to notice in raw tables By reshaping the dataset using melt(), multiple financial variables were converted into a format suitable for visualization, making it easier to compare them within a single chart. This step reinforces an important lesson in analytics: numbers tell the truth, but visuals make the story easier to understand. 📊 Dr James Daniel Paul P Lovely Professional University (LPU) #Python #DataVisualization #Seaborn #BusinessAnalytics #FinanceAnalytics #LearningJourney

📊 Stationarity in Time Series — Practical Python Guide Most forecasting models (like ARIMA) require stationary data. Here’s a quick practical workflow to understand, test, and transform a time series. 🔹 1️⃣ Types of Stationarity • Strong Stationarity: Entire distribution stays the same over time (rare in real data). • Weak Stationarity: • Constant mean • Constant variance • Autocovariance depends only on lag 🔹 2️⃣ Visual Check (Rolling Statistics) 🔹 3️⃣ Statistical Tests 🔹 4️⃣ Making Time Series Stationary •Differencing •Log / Transformation •Detrending Git Repo:- https://lnkd.in/gqtwdXbm 🎯 Key Insight: Before building forecasting models, always test stationarity and transform the series if needed. Grateful to my mentor Ayushi Mishra for guiding me through practical time series concepts. #DataScience #TimeSeries #TimeSeriesAnalysis #Stationarity #ADFTest #KPSSTest #TimeSeriesForecasting #PythonForDataScience #Statistics #StatisticalModeling #MachineLearning #TechCommunity #AnalyticsCommunity #DataScientist #LearningInPublic

𝗠𝗔𝗖𝗛𝗜𝗡𝗘 𝗟𝗘𝗔𝗥𝗡𝗜𝗡𝗚 𝗙𝗢𝗥 𝗕𝗘𝗚𝗜𝗡𝗡𝗘𝗥𝗦 𝗣𝗮𝗻𝗱𝗮𝘀 𝗙𝘂𝗻𝗱𝗮𝗺𝗲𝗻𝘁𝗮𝗹𝘀: 𝗦𝗲𝗿𝗶𝗲𝘀, 𝗗𝗮𝘁𝗮𝗙𝗿𝗮𝗺𝗲𝘀, 𝗜𝗻𝗱𝗲𝘅𝗶𝗻𝗴, 𝗙𝗶𝗹𝘁𝗲𝗿𝗶𝗻𝗴 & 𝗦𝗼𝗿𝘁𝗶𝗻𝗴 Data is everywhere, but extracting meaningful insights from it requires the right tools. Pandas is one of the most powerful libraries in Python for data analysis and manipulation. In this notebook, I covered the core building blocks of Pandas starting from the fundamentals and moving towards practical data operations. From creating Series and DataFrames to performing filtering, indexing, sorting, and ranking, this notebook provides a strong foundation for working with real-world datasets. If you're starting your data science or machine learning journey, mastering Pandas is an essential step. #GenerativeAI #Python #Pandas #DataScience #MachineLearning